Paul Feuerstadt, MD1, Karthik Gnanapandithan, MD, MS2, Olga Aroniadis, MD, MSc3, Sohail Mansoor, MD4, Abdul Bhutta, MD5, Michael Silverman, MD6, Marc Fenster, MD7, Lawrence J. Brandt, MD, MACG7; 1Yale University, School of Medicine, Hamden, CT; 2Mayo Clinic, Jacksonville, FL; 3Stony Brook University School of Medicine, Stony Brook, NY; 4Albany Medical Center, Albany, NY; 5State University of New York Upstate Medical Center, Syracuse, NY; 6Cleveland Clinic Foundation, Cleveland, OH; 7Montefiore Medical Center, Bronx, NY
Introduction: CI is the most common ischemic injury to the GI tract. The clinical utility of antibiotic treatment (ABX) in CI is unclear and the literature is limited. We hypothesized that ABX leads to improved patient outcomes in CI. Methods: We conducted a retrospective cohort study of pts admitted with biopsy-proven CI admitted to Yale-New Haven Hospital and Montefiore Medical Center from 2005-2017. For each pt, demographics, medical co-morbidities, medication exposures, treatments and outcomes were recorded. Pts who received any ABX therapy (CI+ABX) were compared with those that did not (CI-ABX). Subgroup analyses were performed considering the American College of Gastroenterology guidelines for severity of CI (mild/moderate and severe) and according to age (≤65 and >65 yrs). Multivariate logistic regression was used to predict binary outcomes adjusting for the Charlson Comorbidity Index (CCI), severity of CI and presence of concomitant small bowel ischemia (SAS 9.4). Results: 838 pts met inclusion criteria, of whom 4, 407 and 418 had mild, moderate and severe disease, respectively. There were no differences between CI+ABX (n=567, 67.7%) and CI-ABX groups (n=271, 32.3%) in age, gender and CCI. CI+ABX pts had severe CI (54% vs 42%, p=0.001) and required the ICU (26% vs 20%, p=0.04) more frequently than CI-ABX pts. 30-d mortality was 5.8% and 4.1% (p=0.3) in CI+ABX and CI-ABX pts, respectively. 17.4% of CI+ABX pts and 9.7% of CI-ABX pts underwent colectomy (p=0.004). When adjusting for CCI, severity and small bowel involvement, there were no significant differences in 30-d mortality (p=0.93), 30-d colectomy (p=0.15), 90-d recurrence (p=0.38), 90-d readmission (p=0.34) or length of stay (p=0.07) between CI+ABX and CI-ABX groups. When isolating those with mild/moderate disease, severe disease, age ≤ 65 yrs and >65 yrs, there were no differences in the 30-d mortality or colectomy between CI+ABX and CI-ABX. The 4 classes of ABX used most frequently were piperacillin-tazobactam (13.2%), ciprofloxacin-metronidazole (57.1%), ceftriaxone-metronidazole (11.1%) and “other antibiotic” (18.5%). Discussion: ABX does not improve outcomes in pts with CI overall or at any severity level. 30-d mortality was not improved with ABX, even when accounting for age, co-morbidities, severity and small bowel involvement. Given this, clinicians should consider not giving ABX to pts with CI. A prospective, randomized controlled trial, would be important to further validate these results.
Disclosures: Paul Feuerstadt: Ferring/Rebiotix – Consultant. Merck and Co. – Speaker's Bureau. Roche Diagnostics – Consultant. Karthik Gnanapandithan indicated no relevant financial relationships. Olga Aroniadis indicated no relevant financial relationships. Sohail Mansoor indicated no relevant financial relationships. Abdul Bhutta indicated no relevant financial relationships. Michael Silverman indicated no relevant financial relationships. Marc Fenster indicated no relevant financial relationships. Lawrence Brandt indicated no relevant financial relationships.
